View Article Online / Journal Homepage / Table of Contents for this issue

ARTICLE

Development of a microwave-assisted extraction method and isotopic validation of mercury species in soils and sedimentsw
G. M. Mizanur Rahman and H. M. Skip Kingston* Department of Chemistry and Biochemistry, Duquesne University, Pittsburgh, PA 15282, USA. E-mail: kingston@duq.edu; Fax: 412 396 4013 or 412 396 5683; Tel: 412 396 5564
Received 26th March 2004, Accepted 7th January 2005 First published as an Advance Article on the web 4th February 2005

www.rsc.org/jaas

Published on 01 January 2005. Downloaded by Universiteit Utrecht on 28/08/2013 22:10:01.

An ecient and rapid closed vessel microwave-assisted extraction method based on an acidic extractant has been developed to determine inorganic mercury and methylmercury in soils and sediments. Parameters optimized during this study were nitric acid concentration, amount of sample, extraction temperature and irradiation time. The results suggest that the nitric acid concentration and the irradiation temperature are statistically signicant both for extraction eciency and for stability of mercury species. A processed topsoil (Hg o 0.01 ng g1) spiked with inorganic mercury and methylmercury and SRM 2711 (spiked with methylmercury) were used during the method development. The sample preparation was optimized in a closed-vessel system by heating 1.0 g of sample in 10.0 ml of 4.0 mol l1 HNO3 for 10 min at 100 1C with magnetic stirring. Analyses of the extracts were carried out by using three types of instruments, Direct Mercury Analyzer-80 (DMA-80), inductively coupled plasma mass spectrometer (ICP-MS) and highperformance liquid chromatograph coupled with inductively coupled plasma mass spectrometer (HPLC-ICPMS). The results obtained from each of these detection techniques agreed signicantly at the 95% condence level. The method was validated by the analyses of two types of specically prepared reference soil samples and four certied reference materials (BCR 580, SRM 2704, SRM 2709 and SRM 1941a). The inorganic mercury and methylmercury concentrations found were in good agreement at the 95% condence level with the certied or made-to value. The method was also validated using EPA Method 6800 as a diagnostic tool to check whether interconversion of inorganic mercury to methylmercury or vice versa took place during or after extraction; the amount of such interconversions was found to be statistically negligible. The method is in the process of consideration and adopted by the United States Environmental Protection Agency (US EPA) as a primary mercury species extraction protocol from soils and sediments in EPA draft Method 3200.

Introduction
Mercury speciation has long been a eld of concern. Such interest is mainly due to toxicological impact, ecological problems and biogeochemical cycling of mercury involving distribution, accumulation, transformations and transport pathways in the natural environment.1 Mercury is a very toxic element. However, the toxicity of mercury is highly dependent on its chemical form. Methylmercury is one of the most toxic mercury species. To understand the toxicological impact and pathways of mercury species in the environment, the determination of total mercury is frequently not sucient. Therefore, the assessment of inorganic mercury and methylmercury concentrations, specically in sediments and soils, is very important to the interpretation of biogeochemical cycles of mercury in aquatic environments.2 Determination of dierent mercury species from various complex matrices, e.g., soils and sediments, is still considered a dicult task due to the frequently very low concentration of methylmercury in soils and sediments (less than 1.5% of the total mercury).3 The quality of the results mainly depends on the sample pre-treatment stages (sampling, storage and sample preparation), in spite of signicant improvements in the instrumentation techniques. The most widely used methods for the extraction and separation of inorganic and methylmercury are the Westo o technique47 (acidic leaching method), alkaline digestion,810 steam distillation,911 solvent extraction,1214 a
w Electronic supplementary information (ESI) available: optimization of HNO3 concentration, optimization of sample weight, optimization of irradiation temperature and optimization of irradiation time. See http://www.rsc.org/suppdata/ja/b4/b404581e/

modied Westo o methodology15 (alkaline based technique), and supercritical uid extraction,16 followed by one or two separation steps. The separation and detection techniques associated with these methods include gas chromatography (GC), HPLC coupled with element-selective detection techniques, such as ICP-MS, atomic emission spectrometry (AES), atomic absorption spectrometry (AAS), atomic uorescence spectrometry (AFS), or cold vapor atomic absorption spectrometry (CV-AAS). As all of the aforementioned sample preparation methods use either acid or base with organic solvents, and, after extraction, most of them implement sample preconcentration steps (e.g., ethylation or reduction with SnCl2 or hydride generation with NaBH4), there is a possibility of interconversion or unidirectional transformation of inorganic mercury to organic mercury13,1629 or vice versa2932 during sample storage, shipment, extraction, pre-concentration or analysis steps. Therefore, the results obtained using these procedures frequently introduce positive or negative biases for either inorganic mercury or methylmercury, or both. Besides such drawbacks, these methods require much solvent, labor and time. The eciency of the less solvent- and time-consuming microwave-assisted extraction (MAE) technique for sample preparation in environmental applications has been evaluated elsewhere in dierent matrices (soils, sediments, and biological tissues) in dierent applications (total digestion for elemental analysis, extraction of selected organic compounds), and in speciation analysis (organotin). Vazquez et al.33,34 used the focused microwave-assisted extraction (FMAE) technique to extract methylmercury with HCl and toluene, a modied method of Westo o ,4,5 from sediment and biological tissue samples. Tseng et al.3537 also implemented FMAE for the
J. Anal. At. Spectrom., 2005, 20, 183191 183

DOI: 10.1039/b404581e

This journal is & The Royal Society of Chemistry 2005

View Article Online

extraction of methylmercury, also from sediment and tissue samples. There are several drawbacks to FMAE: samples must be extracted at atmospheric pressure and below the boiling point of the solvent; simultaneous extraction of multiple samples is not possible; it is dicult to preset a constant temperature prole as this technique only allows control of the applied power which, in turn, is directly dependent on the number of samples or the total mass, and, there is a high possibility of losing the volatile organomercury compounds during extraction. However, no one has yet tried the closedvessel microwave-assisted extraction technique (which is free from the aforementioned drawbacks) for mercury speciation in soils or sediments. Therefore, the purpose of this study was to develop a microwave-assisted extraction procedure capable of quantitative extraction with little or no transformation of inorganic mercury and methylmercury from soils and sediments in a closed-vessel microwave system, and incorporate it into EPA draft Method 3200 as an alternative extraction procedure for mercury species. Careful optimization of the conditions for the microwave extraction procedure is required to stabilize the mercury species in the microwave eld, prior to speciation analysis. Essential parameters, such as concentration of the extraction medium, amount of sample, temperature and time of exposure must be optimizedw. The literature35 suggests that nitric acid (HNO3) is a better solvent for microwave-assisted extraction because it introduces little or no interferences to the ICP-MS. Therefore, nitric acid has been evaluated as an extraction solvent. The irradiation power, one of the most useful parameters for microwave extraction, was not optimized during this study due to its dependency on the number of samples or the total mass of the extraction medium. This paper describes a fast and easy method for the quantitation of inorganic mercury and methylmercury using closedvessel microwave-assisted extraction, followed by separation with HPLC and detection with ICP-MS. The stability of the mercury species in a microwave eld and the optimization of dierent parameters are also described in detail. The developed method was then validated by using dierent standard reference materials and reference soils obtained from Environmental Resource Associatess. The developed method was also validated using EPA Method 6800 [Elemental and Speciated Isotope Dilution Mass Spectrometry, (IDMS and SIDMS, respectively)].38 EPA Method 6800 was used as a diagnostic tool to check whether any interconversion between inorganic mercury and methylmercury is taking place during or after extraction. One of the unique applications of SIDMS is to trap errors related to specic portions of a protocol. This is accomplished by using multiple spikings with multiple isotope-labeled species at specic method protocol points. The error of the specic steps can be discovered, and their contribution to the overall transformation of a species known. To perform these types of applications, inorganic mercury and methylmercury labeled with multiple isotopes are required. Although methylmercury labeled with dierent isotopes has recently become commercially available,3941 the methylmercury labeled with a minor mercury isotope was synthesized in the laboratory.

anhydride, Na2S2O3, toluene, sodium acetate, ammonium acetate, 2-mercaptoethanol (98%), and optima grade methanol were obtained from Fisher Scientic (Pittsburgh, PA, USA). Mercaptoacetic acid (97%) was obtained from Aldrich (Milwaukee, WI, USA). The reagent grade tetramethyltin (98%) was obtained from Alfa Aesar (Ward Hill, MA, USA). Standard solutions and certied reference materials A standard stock solution of 1000 mg ml1 of HgCl2 in 5% HNO3 and 1000 mg ml1 of CH3HgCl in water were commercially available from Alfa Aesar (Ward Hill, MA, USA). All stock solutions were stored in amber glass bottles in a cold room at 4 1C. Working standards were prepared daily by proper dilution with DDI water. 201 HgO and 199HgO were obtained from Isotech Inc. (Miamisburg, OH, USA). Approximately 100 mg g1 of stock 199 Hg21 spike was prepared by dissolving B10 mg of 199HgO in 2 ml concentrated HCl and made up to 90 g with 1% HNO3. CH3201Hg1 was synthesized from 201HgO using tetramethyltin.39 To prepare 201HgCl2, 6 ml of 201Hg21 solution (11 mg ml1) was mixed with 2 ml of 6.0 mol l1 HCl in a 20 ml amber glass vial and stirred for 5 min. A 0.93 mol l1 methanolic solution of (CH3)4Sn was prepared by dissolving 0.340 g of (CH3)4Sn in 2 ml of methanol. This reagent was quantitatively transferred into the 201HgCl2 solution and the glass vial cap was put back on. The resulting reaction mixture was then stirred for 1 h in a 60 1C water bath to complete the synthesis. The reaction mixture was cooled to room temperature and extracted 3 times with toluene (4 3 3 ml). The synthesized methylmercury (in toluene) was washed with DDI water 3 times (4 3 3 ml). The toluene extract was then extracted twice with 2.5 ml of 1% Na2S2O3. All of the extracts were stored in amber glass vials in a cold room at 4 1C. The nal concentrations of the stock CH3201Hg1 in 1% Na2S2O3 and 199 Hg21 in 1% HNO3 were determined by reverse isotope dilution calibration. Working standards of CH3201Hg1 and 199 Hg21 were prepared daily by diluting with DDI water. Caution. Mercury compounds, especially methylmercury, are highly toxic materials. Proper knowledge and safety guidelines regarding working with mercury compounds are required to handle these compounds. The sulfhydryl cotton ber (SCF) as a solid phase extraction (SPE) medium was prepared according to Han et al.12 A mixture of reagent containing 50 ml mercaptoacetic acid, 35 ml acetic anhydride, 16 ml acetic acid, 0.15 ml H2SO4 and 5 ml DDI water was prepared in a clean beaker. A 15 g portion of cotton was immersed in this reagent mixture, and kept it in an oven at 40 2 1C for 4 d. The product was removed from the oven and washed with DDI water on a vacuum lter funnel until the pH of the washing was neutral. The cotton product (now SCF) was then dried in an oven at 40 2 1C for 2 d. SCF Eluent 1 (1 mol l1 HCl 1 mol l1 NaCl) for methylmercury was prepared by diluting 20.7 ml of concentrated HCl to 200 ml in DDI water, then dissolving 14.6 g of NaCl in the prepared solution and making up to 250 ml with DDI water. SCF Eluent 2 (6 mol l1 HCl saturated NaCl 0.1% CuCl2 2H2O) for inorganic mercury was prepared by diluting 124 ml of concentrated HCl to 200 ml in DDI water, then dissolving 11 g of NaCl and 0.25 g of CuCl2 2H2O in the prepared solution and making up to 250 ml with DDI water. A 0.2 mol l1 acetate buer (pH 3.6) was prepared by dissolving 1.14 ml of glacial acetic acid and 0.1148 g of sodium acetate in 100 ml DDI water. HPLC speciation mobile phase [30% (v/v) methanol 0.06 mol l1 ammonium acetate 0.005% 2-mercaptoethanol], modied from Wilkens procedure,42 was prepared by diluting 300 ml of methanol, 50 ml of 2-mercaptoethanol and 4.8 g ammonium acetate with 700 ml of DDI water.

Published on 01 January 2005. Downloaded by Universiteit Utrecht on 28/08/2013 22:10:01.

Experimental
Reagents and chemicals Analytical grade nitric acid (Fisher Scientic, Pittsburgh, PA, USA) and double de-ionized (DDI) water (18 MO cm1), prepared from a Barnstead NANOpure Ultrapure Water System (Dubuque, IA, USA), were used. Dierent concentrations of nitric acid were prepared by diluting an appropriate volume of nitric acid in DDI water. Reagent grade HCl, H2SO4, NaCl, CuCl2 2H2O, NaOH, acetic acid, acetic
184 J. Anal. At. Spectrom., 2005, 20, 183191

View Article Online

NIST SRMs 1941a (Organics in Marine Sediment, 0.5 0.2 mg Hg kg1), 2704 (Bualo River Sediment, 1.44 0.07 mg Hg kg1), 2709 (San Joaquin Soil, 1.40 0.08 mg Hg kg1), and 2711 (Montana Soil, 6.25 0.19 mg Hg kg1), IRMM reference material BCR 580 (Estuarine Sediment, 132 3 mg Hg kg1 and 75 4 ng CH3Hg1 g1), topsoil (100% processed topsoil, mercury content, o0.01 ng Hg g1) and reference soils (Material-1: 100% processed topsoil, and Material-2: 75% processed topsoil and 25% Ottawa Sand) from Environmental Resource Associatess (ERA) (Arvada, CO, USA) were used for method development and validation. Instrumentation A laboratory microwave system (Ethos 1600) (Milestone, Monroe, CT, USA), equipped with temperature and pressure feedback control and a magnetic stirring capability, was used in this study. This device is accurate in sensing temperature within 2.0 1C of set temperature, and automatically adjusts the microwave eld output power. This device extracts ten samples simultaneously. The high pressure closed digestion vessels used for extraction are made of high purity TFM (a thermally resistant form of Teon) and have a capacity of 100 ml. Caution. Safety guidelines regarding work with microwave elds in the laboratory must be observed.43 A ConstaMetric 4100Bio/MS polymeric inert pump (Thermo Separation Products, Riviera Beach, FL, USA) and a 5 mm Supelcosil LC-18 HPLC column with a Pelliguard LC-18 guard column (Supelco, Bellefonte, PA, USA) were used in this study to separate inorganic mercury and methylmercury. A six-port injection valve (Valco Instrument Co. Inc., Houston, TX, USA) was placed between the pump and the column. Because no special interface is required between the LC-18 column and the ICP-MS, one outlet of the column is directly interfaced with the nebulizer of the ICP-MS using a piece of TFM tubing; the other end is connected to a 50 ml TEFZELt sample loop (CETAC Technologies, Omaha, NE, USA). The mobile phase was buered 30% methanol (refer to Standard Solutions and Certied Reference Materials). The SPE apparatus used for this study was an SPE manifold (Supelco, Bellefonte, PA, USA) with the custom SCF SPE column prepared according to Han et al.12 An HP 4500 ICP-MS (Agilent Technologies, Palo Alto, CA, USA and Yokogawa Analytical Systems Inc., Tokyo, Japan) was used as the detector for the HPLC system in this study. The HP 4500 ICP-MS was also used for direct determination of total mercury from sample extracts. The sample delivery system consisted of a peristaltic pump and a Scott-type double pass quartz spray chamber with concentric nebulizer and quartz torch. The instrument was tted with platinum sampler and skimmer cones, and optimized daily using 10 ppb tuning solution (Agilent Technologies, Palo Alto, CA, USA) containing Li, Y, Ce and Tl in 30% methanol (for speciation analysis), and in 2% HNO3 (for total mercury analysis). The time resolved analysis (TRA) mode was engaged for speciation analyses, and the spectrum mode was engaged for total mercury analyses. ICP-MS operating conditions for total mercury and speciation measurements are given in Table 1. A Direct Mercury Analyzer-80 (DMA-80; Milestone, Monroe, CT, USA) was used in this study to determine the total mercury content in each of the extraction and purication steps. The operation for the DMA-80 used throughout this work was based on the guidelines provided in EPA Method 7473 protocol.44,45 The DMA-80 operates on the basis of thermal decomposition, catalytic reduction, amalgamation, desorption and atomic absorption spectrometry. Prior to decomposition, the sample is initially dried in an oxygen stream passing through a quartz tube located inside a controlled heating coil. The combustion gases are further decomposed on a catalytic column at 750 1C and mercury vapour is
Published on 01 January 2005. Downloaded by Universiteit Utrecht on 28/08/2013 22:10:01.

Table 1 Operating conditions for ICP-MS Plasma conditions Radiofrequency power/W Plasma gas ow rate/l min1 Auxiliary gas ow rate/l min1 Measurement parameters Acquisition mode Isotope monitored Nebulizer gas ow rate/l min1 Liquid sample ow rate/ml min1 Integration time per point/s Total analysis time/s Replicates
a

1450 15.0 1.0 Spectruma, Time resolved analysis (TRA)b 199 Hg, 200Hg, 201Hg, 202Hg 1.0a, 0.93b 1.0ab 0.5 65a, 450b 5a, 1b

Total Hg analysis.

Hg speciation analysis.

collected on a gold amalgamation trap and subsequently desorbed for quantitation. Mercury content is determined using atomic absorption spectrometry at 254 nm. Preparation of SCF column, preconcentration and mercury species separation The SCF column was prepared in a 5 ml disposable hypodermic syringe (Aldrich, Milwaukee, WI, USA). A PTFE frit (Supelco, Bellefonte, PA, USA) was added at the bottom of the syringe, and then a 0.2 g portion of SCF along with 3 ml DDI water was added. A second frit was placed on top of the water and the syringe was shaken for 23 min to get a homogenized mixture. Pressure was applied on the top frit to compact the SCF into a homogeneous disk between the two frits. The SCF columns were conditioned just before application by passing 10 ml DDI water, then 10 ml 6 mol l1 HCl and nally 15 ml DDI water with a ow rate of not more than 1 ml min1. The pH values of the extracts were adjusted to 6 1 with 10 mol l1 NaOH and they were ltered through a 1 mm lter to retain particles larger than 1 mm. The retained particles were rinsed with 0.1 mol l1 HCl. The ltered solutions and the rinsed solutions were combined. 1 ml of 0.2 mol l1 acetate buer was added to each solution and the pH was re-adjusted to 34 by adding 6 mol l1 HCl. The solutions were then passed through the conditioned SCF column at a ow rate r1 ml min1. Both mercury species were retained in the SCF medium. Methylmercury was eluted from the SCF medium by passing 8 ml of SCF Eluent 1 through the medium followed by 2 ml of DDI water at a ow rate r1 ml min1. The inorganic mercury is still retained in the SCF medium. Inorganic mercury was then eluted from the SCF medium by passing 8 ml of SCF Eluent 2 through the medium followed by 2 ml DDI water at a ow rate r 1 ml min1. Optimized analytical procedure Samples of approximately 1.0 g of homogenized soil or sediment and 10 ml of 4 mol l1 HNO3 were placed in the microwave extraction vessels. A magnetic stirrer bar was added to each vessel for thorough mixing of solvent with the sample. Microwave vessels were sealed and irradiated at 100 1C for 10 min with magnetic stirring taking place. A 2 min ramping time was used to reach the desired temperature of 100 1C. After microwave irradiation, the vessels were cooled to room temperature, ltered through a 0.22 mm glass ber lter (Fisher Scientic, Pittsburgh, PA, USA) and stored in the cold room until analyzed (usually less than 2 d). Blanks were prepared along with the samples in each batch. To evaluate the stability of mercury species in the microwave eld, 10 ml nitric acid solutions at dierent concentrations
J. Anal. At. Spectrom., 2005, 20, 183191 185

View Article Online

were spiked with 100 ml of Hg standard (100 mg Hg ml standard) and 100 ml of CH3Hg1 standard (100 mg CH3Hg1 ml1 standard) and irradiated at dierent temperatures and for dierent irradiation times. The topsoil was spiked with the same concentrations of inorganic mercury and methylmercury; and the SRM 2711 was spiked only with methylmercury, to which the extraction solvent was added in order to optimize the microwave assisted extraction procedure. The samples were then irradiated by varying both irradiation time and irradiation temperature.

21

21

Sample preparation for interconversion study

Published on 01 January 2005. Downloaded by Universiteit Utrecht on 28/08/2013 22:10:01.

mixed mercury standard. It was found that both mercury species were stable within the time range studied (Table 2). After careful evaluation of all data sets (Table 2), it was concluded that mercury species are stable in 4 mol l1 nitric acid at temperatures up to 100 1C and for at least 14 min of microwave irradiation. Results may vary due to temperature and time eects, or show dierent trends with a higher concentration of nitric acid. During this study, only the extraction solvent was spiked with mixed mercury standard. Results may also vary or show dierent trends with soil or sediment samples. Therefore, the next step was to use soil and sediment samples or SRMs to develop a methodology for the microwave-assisted extraction of mercury species. Optimization of HNO3 concentration The nitric acid concentration eects on the extraction eciency and stability of mercury species in soils and SRM 2711 were studied. Approximately 0.4 g of each soil sample and SRM 2711 were weighed directly in the microwave vessel. Samples were left for 1 h to equilibrate, and then 10 ml of extraction solvent (1.0, 2.5, 4.0, 5.5 and 7.0 mol l1 HNO3) was added into the microwave vessel. The sample was then extracted at 50 1C with the following microwave procedure. Step 1: Time 2 min (Ramping to 50 1C); Temperature 50 1C; Power 1000 W. Step 2: Time 5 min (Hold at 50 1C); Temperature 50 1C; Power 1000 W. Note: Automated feedback control was engaged for both protocol steps; Venting Time 3 min. After each extraction cycle was completed, the samples were cooled to room temperature and ltered through 0.22 mm glass ber lter and stored in the cold room at 4 1C until analyzed. The extracts were analyzed with the DMA-80 and the ICP-MS for total mercury concentration, by direct aspiration in spectrum mode, and with the HPLC-ICP-MS for inorganic mercury and methylmercury concentrations. In order to validate the quantication ability of the three individual detection techniques, the total mercury values for the HPLC-ICP-MS technique were determined from the summation of inorganic mercury and methylmercury concentration values. Results from dierent analysis methods were statistically indistinguishable at 95% CL (Table 3). It was found that 95 7% to 107 8% of total mercury is extractable from spiked topsoil using 1.0 to 7.0 mol l1 HNO3 (DMA-80). On the other hand, the SRM 2711 extraction eciency is highly dependent on the concentration of the solvent used. The extraction eciency increases from 54 8% to 92 5% by increasing the nitric acid concentration from 1.0 mol l1 to 7.0 mol l1 (DMA-80). It is evident that the sample matrix inuences the extraction eciency. In the case of topsoil, the spikes were freshly added and had very limited time to interact physically and/or chemi-

About 1 g of sediment or soil samples along with magnetic stirrer bar was placed into microwave vessels. Samples were double spiked with known amounts of dilute aqueous solutions of isotopically enriched CH3201Hg1 and 199Hg21 standard as SIDMS internal standard and artifact formation controller, in such a way that the desired isotope ratio became approximately 1 : 1, and left for 1 h for equilibration. The exact amounts of isotopic spike depend on the levels of inorganic mercury and methylmercury present in the samples and the spike concentration. 10 ml of 4.0 mol l1 HNO3 was added in each vessel and extracted according to the MAE procedure described in Optimized analytical procedure.

Results and discussion

Stability of mercury species under microwave irradiation The eects of nitric acid concentration on the stability of mercury species was studied at dierent HNO3 concentrations (0, 0.5, 1.0, 1.5, 2.0, 2.5, 3.0, 3.5 and 4.0 mol l1 HNO3) at 50 1C for 5 min. A mixed mercury standard (100 mg ml1) was used for spiking. 10 ml of extraction solvent for each concentration level was measured and dispensed into a series of microwave vessels. 100 ml of the mixed standard was spiked into each of the vessels and a magnetic stirring bar was added. The samples were irradiated for 5 min and analyzed with HPLC-ICP-MS. It was found that both mercury species were stable at that temperature for 5 min up to 4.0 mol l1 HNO3 concentrations (Table 2). The temperature eect on the stability of mercury species was studied at dierent irradiation temperatures (30, 40, 50, 60, 70, 80, 90 and 100 1C) using 4.0 mol l1 HNO3 (spiked with 100 ml of mixed mercury standard) for 5 min and analyzed with HPLC-ICP-MS. It was found that both mercury species were stable within this temperature range (Table 2). The irradiation time eect on the stability of mercury species was also studied at dierent irradiation times (2, 4, 6, 8, 10, 12 and 14 min) using 4.0 mol l1 HNO3 and irradiated at 100 1C. The solvent was also spiked with the same concentration of
Table 2

Eects of microwave irradiation on mercury species: the results are expressed in percent recovery Temperature eectsa (4.0 mol l1 HNO3 and 5 min) CH3Hg1 4 4 5 8 8 8 6 3 6 96 97 94 93 99 89 89 95 97 4 4 8 11 8 8 7 2 8 (1C) 30 40 50 60 70 80 90 100 Hg21 93 88 92 96 96 101 97 103 6 6 14 11 14 10 4 9 CH3Hg1 93 106 98 101 96 103 103 99 13 8 7 7 13 8 4 8 Time eectsa (4.0 mol l1 HNO3 at 100 1C) (min) 2 4 6 8 10 12 14 Hg21 93 100 97 100 89 89 91 12 14 5 2 15 14 15 CH3Hg1 96 103 107 91 100 99 89 17 13 9 5 6 3 8

HNO3 concentration eectsa (at 50 1C for 5 min) (mol l1) 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0
a

Hg21 101 97 90 90 92 91 93 89 97

Uncertainties are expressed at 95% CL, n 3.

186

J. Anal. At. Spectrom., 2005, 20, 183191

View Article Online

Table 3 Percent recovery results for optimization of HNO3 concentration HPLC-ICP-MSa (%) Sample Topsoil (Spiked) HNO3 concentration (mol l1) 1.0 2.5 4.0 5.5 7.0 1.0 2.5 4.0 5.5 7.0 DMA-80 (total Hg) (%) 95 105 103 100 107 54 57 70 74 92 7 2 2 2 8 8 5 1 2 5
a

ICP-MS (total Hg) (%) 98 101 95 94 102 57 61 68 77 88

b

Hg21 101 106 103 111 128 44 51 74 103 147 7 1 1 7 5 8 7 8 7 19

CH3Hg1 91 107 105 87 69 81 77 83 59 47 6 11 5 17 3 3 9 5 8 17

Total Hgb 96 106 104 99 99 63 64 79 81 97 5 6 3 9 3 4 6 5 5 13

SRM 2711 (Spiked)

Published on 01 January 2005. Downloaded by Universiteit Utrecht on 28/08/2013 22:10:01.

2 3 2 2 2 2 2 3 2 1

a Uncertainties are reported as 95% CL with n 4 (no. of individual sample). MS.

Total Hg average of Hg21 and CH3Hg1 values from HPLC-ICP-

cally with soil particles, and were easy to extract with solvents at dierent concentrations. On the other hand, SRM 2711 is a natural soil and inorganic mercury is naturally tightly bound with the soil particles. Therefore, it was dicult to extract with less concentrated extraction solvents. However, from the speciation data for concentration eect on extraction eciency and stability of mercury species (Table 3), it is observed that the percent recovery for inorganic mercury increased from 101 7% at 1.0 mol l1 HNO3 to 128 5% at 7.0 mol l1 HNO3 for spiked topsoil and from 44 8% at 1.0 mol l1 HNO3 to 147 19% at 7.0 mol l1 HNO3 for SRM 2711 (spiked). On the other hand, the percent recovery for methylmercury increased from 91 6% at 1.0 mol l1 HNO3 to 105 5% at 4.0 mol l1 HNO3 for spiked topsoil and then gradually decreased to 69 3% at 7.0 mol l1 HNO3. For SRM 2711 (spiked with methylmercury), the trend was dierent. In this case the percent recovery of methylmercury was almost stable up to 4.0 mol l1 HNO3 and then decreased gradually to 47 17% at 7.0 mol l1 HNO3. The degradation of methylmercury leads to decreasing recoveries for methylmercury and apparent increasing recoveries for inorganic mercury. As the methylmercury was stable up to 4.0 mol l1 HNO3 and degraded at higher concentrations for both evaluated materials, the 4.0 mol l1 HNO3 was used as optimized extraction solvent concentration throughout the study. Optimization of sample weight The eect of sample weight on the extraction eciency was studied using the same topsoil and SRM 2711. Dierent amounts (0.25, 0.50, 0.75, 1.00 and 2.00 g) of topsoil and SRM 2711 were weighed and processed as described in the
Table 4 Percent recovery results for optimization of sample weight

Experimental section. Extracts were analyzed using three dierent instruments. The results are summarized in Table 4 and are statistically indistinguishable at 95% CL. From Table 4, the total mercury results obtained from spiked topsoil, it is found that the recovery for 0.25 g of sample was 108 5% and in the range of 90 4% to 97 4% for all the other sample amounts studied (DMA-80): statistically, there was no signicant dierence between the last set of results. But, on the other hand, the recovery from SRM 2711 (spiked with methylmercury) was 65 4% for 0.25 g and 59 4% for 0.50 g, which are statistically indistinguishable at their 95% CL. The percent recoveries for all the other sample amounts studied were in the range of 51 2% to 57 3%. The speciation data (Table 4) indicates that the sample weight has very little or no eect on the extraction eciency at 50 1C. In case of spiked topsoil, the percent recovery for inorganic mercury was from 91 3% to 102 7% and for methylmercury was from 97 3% to 106 7%. But in SRM 2711, the recovery of inorganic mercury was poor (from 33 2 to 47 3%). On the other hand, the recovery of spiked methylmercury for SRM 2711 was from 73 1 to 92 5% and was stable. The robustness of the extraction method is demonstrated in this study by optimizing the sample amount over one order of magnitude. For the entire sample range tested, statistically identical recoveries were obtained from 0.25 g to a 2.0 g aliquot of sample. An intermediate 1.00 g sample size was used as optimized sample amount during rest of the evaluations. Optimization of irradiation temperature The eect of irradiation temperature on the extraction eciency and stability of the mercury species was studied using spiked

HPLC-ICP-MSa (%) Sample Topsoil (Spiked) Sample weight/g 0.25 0.50 0.75 1.00 2.00 0.25 0.50 0.75 1.00 2.00 DMA-80a (total Hg) (%) 108 97 92 90 94 65 59 57 51 55
b

ICP-MSa (total Hg) (%) 103 98 96 94 98 61 57 60 49 57 1 1 1 1 1 1 1 1 1 1

Hg21 102 100 91 92 100 47 34 34 33 34 7 8 3 9 9 3 4 1 2 2

CH3Hg1 97 106 98 106 106 74 85 92 73 87 3 5 1 7 4 1 2 5 1 2

Total Hgb 100 103 95 99 103 61 60 63 53 61 4 5 2 6 5 2 2 3 1 1

SRM 2711 (Spiked)

5 4 6 4 9 4 4 3 2 2

Uncertainties are reported as 95% CL with n 4.

Total Hg average of Hg21 and CH3Hg1 values from HPLC-ICP-MS.

J. Anal. At. Spectrom., 2005, 20, 183191

187

View Article Online

Table 5 Percent recovery results for optimization of irradiation temperature HPLC-ICP-MSa (%) Sample Topsoil (Spiked) Irradiation temperature/1C 50 60 70 80 90 100 110 120 130 50 60 70 80 90 100 110 120 130 DMA-80 (total Hg) (%) 90 98 96 97 103 103 103 103 105 51 57 64 71 82 93 94 92 97 4 2 1 1 5 6 2 6 2 2 1 1 2 4 7 7 4 4
b a

ICP-MS (total Hg) (%) 94 99 101 95 97 97 101 100 97 49 56 66 72 78 91 96 97 99 1 2 2 2 1 2 1 1 1 1 1 2 1 2 2 1 1 1

Hg21 92 91 91 88 94 95 94 102 98 33 31 48 69 76 98 109 112 116 9 10 9 6 8 2 4 10 10 2 3 1 5 8 5 6 5 10

CH3Hg1 106 98 102 99 105 101 99 98 96 73 80 95 95 107 102 96 70 79 7 2 4 9 8 8 5 9 4 1 8 6 7 9 7 7 7 11

Total Hgb 99 95 97 94 100 98 97 100 97 53 56 72 82 92 100 103 91 98 6 5 5 5 6 4 3 7 5 1 4 3 4 6 4 5 4 7

Published on 01 January 2005. Downloaded by Universiteit Utrecht on 28/08/2013 22:10:01.

SRM 2711 (Spiked)

Uncertainties are reported as 95% CL with n 4.

Total Hg average of Hg21 and CH3Hg1 values from HPLC-ICP-MS.

topsoil and SRM 2711 (spiked with methylmercury) at dierent temperatures (50, 60, 70, 80, 90, 100, 110, 120 and 130 1C). Samples were processed as described in the Experimental section. The extracts were analyzed with three dierent instruments and results were statistically indistinguishable at 95% CL (Table 5). It was found from Table 5 that the recoveries of the total mercury in spiked topsoil were from 90 4% at 50 1C to 105 2% at 130 1C, and are statistically indistinguishable at 95% CL. On the other hand, the recoveries for total mercury in SRM 2711 (spiked with methylmercury) increased from 51 2% at 50 1C to 97 4% at 130 1C (DMA-80). The speciation data are shown in Table 5. It was found that the recovery results for inorganic mercury in spiked topsoil were from 92 9% at 50 1C to 98 10% at 130 1C, and for methylmercury were from 106 7% at 50 1C to 96 4% at 130 1C, and are statistically indistinguishable at 95% CL throughout the temperature range studied. However, the extraction eciency of inorganic mercury in SRM 2711 increased from 33 2% at 50 1C to 116 10% at 130 1C. The recovery for methylmercury in SRM 2711 (spiked with methylmercury) also increased from 73 1% at 50 1C to 102 7% at 100 1C, then decreased to 79
Table 6 Percent recovery results for optimization of irradiation time

11% at 130 1C. The simultaneously increasing recoveries of Hg21 suggest a degradation of methylmercury at higher temperature. As the recovery for both inorganic mercury and methylmercury was 98 4% and 102 7%, respectively, at 100 1C, this temperature was used as optimum throughout the study. Optimization of irradiation time The eect of irradiation time on the extraction eciency and stability of mercury species was studied at dierent irradiation times (5, 10, 15, 20, 25 and 30 min). Samples were processed as described in the Experimental section. The extracts were analyzed with same three instruments. The results are summarized in Table 6 and are statistically indistinguishable at their 95% CL. From the nal mercury results, the recovery of total mercury in spiked topsoil was from 103 6% at 5 min to 107 2% at 30 min, and in SRM 2711 (spiked with methylmercury) was from 93 7% at 5 min to 99 5% at 30 min (DMA-80) and were statistically indistinguishable at 95% CL throughout the studied time periods. The speciation data for

HPLC-ICP-MSa Sample Topsoil (Spiked) Irradiation time/min 5 10 15 20 25 30 5 10 15 20 25 30

a

DMA-80a (total Hg) (%) 103 107 106 105 104 107 93 91 94 95 93 99 6 7 8 5 6 2 7 5 5 4 5 5
b

ICP-MSa (total Hg) (%) 97 98 102 104 106 104 91 99 97 102 100 100 2 3 1 2 2 2 2 1 5 2 5 2

Hg21 95 91 92 94 91 91 98 100 93 93 94 103 2 9 9 4 10 10 5 9 10 8 10 9

CH3Hg1 101 94 93 95 101 95 102 97 94 93 90 88 8 8 9 11 5 3 7 10 4 13 11 8

Total Hgb 98 93 93 95 96 93 100 99 94 93 92 96 4 6 6 6 6 5 4 7 5 8 7 6

SRM 2711 (Spiked)

Uncertainties are reported as 95% CL with n 4.

Total Hg average of Hg21 and CH3Hg1 values from HPLC-ICP-MS.

188

J. Anal. At. Spectrom., 2005, 20, 183191

View Article Online

Published on 01 January 2005. Downloaded by Universiteit Utrecht on 28/08/2013 22:10:01.

both spiked blank soil and SRM 2711 (spiked with methylmercury) are shown in Table 6. From the speciation data, it was also found that the recovery results for inorganic mercury was from 95 2% at 5 min to 91 10% at 30 min, and for methylmercury was from 101 8% at 5 min to 95 3% at 30 min in the spiked topsoil with no distinguishable degradation of methylmercury during the studied time period. On the other hand, the recovery of inorganic mercury was from 98 5% at 5 min to 103 9% at 30 min, and for methylmercury was from 102 7% at 5 min to 88 8% at 30 min in SRM 2711 (spiked with methylmercury) and results were stable up to 25 min (90 11%), after which degradation of methylmercury is suspected. As a result, the recovery of inorganic mercury increased and methylmercury recovery decreased, although at 95% CL, these changes in recovery were not statistically distinguishable. In order to shorten the sample preparation time, it was decided to use 10 min as the optimum time for extraction. The venting time used throughout this study was 3 min. The cooling rate of the vessels depends on the make, model and type of both the microwave and the vessels used. Therefore, the recommended venting or cooling time used may be more than 3 min.

Validation of the developed and optimized method using reference soil and SRMs The microwave-assisted extraction method was validated by using two dierent sets of reference soil samples (Lot # 061101-02), prepared by Environmental Resource Associatess for SAIC and the United States Environmental Protection Agency (US EPA) from 100% processed topsoil and a mixture of 75% topsoil and 25% Ottawa sand. The preparation of these reference soil samples is described elsewhere.46 Both of these materials contain Hg21 (inorganic mercury), CH3Hg1 (organic mercury) and an equal mixture of Hg21 and CH3Hg1 (mixed mercury). SRM 1941a, SRM 2704, SRM 2709 (certied for total mercury), and BCR 580 (certied for both total mercury and methylmercury) were used for method validation. All of the reference soil samples and NIST SRMs were analyzed directly (as solid, without any sample preparation) with the DMA-80 using EPA Method 7473 protocol before extraction by the microwave-assisted extraction method. Results are summarized in Table 7. The results obtained for dierent soil samples and SRMs from the direct mercury analyses, except for organic mercury in Material-1, are indistinguishable from their corresponding certied-value at 95% CL. Samples and NIST SRMs were then processed according to the optimized MAE procedure described in the Experimental section. The extracts were analyzed with DMA-80 and ICP-MS for total mercury and

with HPLC-ICP-MS for mercury speciation and quantication. As the results obtained from dierent instrumental analyses overlapped at 95% CL and were statistically indistinguishable, only the speciation results for dierent samples are summarized in Table 7. The results demonstrated that good recoveries of methylmercury and inorganic mercury were achieved for all the studied materials and were in excellent agreement with the certied or known values at 95% CL. The 100% recoveries of both mercury species indicates no conversion of species during extraction. The IRMM reference material BCR 580 was also processed according to the MAE procedure described in the Experimental section, and extracts were analyzed with DMA-80 and ICPMS for total mercury measurement (Table 8). Although the amount of methylmercury present in BCR 580 was higher than the HPLC-ICP-MS detection limit (5 ng g1), it could not be analyzed directly using this instrument for speciation. The ratio of methylmercury to inorganic mercury in BCR 580 is 1 : 1760. When the HPLC-ICP-MS was used for speciation, the methylmercury peak was overlapped with the inorganic mercury peak and could not be quantied. This reveals the limitations of the HPLC separation technique for extracts containing analytes at a very high concentration dierence with close retention times. This would not happen if a GC-MS or a GC-ICP-MS were used for speciation and quantication. Therefore, the extracts were speciated rst using SCF SPE methodology (refer to Preparation of SCF column, preconcentration, and mercury species separation), and then analyzed and quantied with HPLC-ICP-MS. The results are summarized in Table 8. The amount of methylmercury and inorganic mercury determined by the SCF-SPE-HPLC-ICP-MS procedure is 0.073 0.002 mg g1 and 133 6 mg g1, which results are in excellent agreement with their corresponding certied values at 95% CL. From the speciation results, it is observed that the MAE method is highly ecient in extracting dierent mercury species from the variety of matrices tested during this study with 99100% recovery.

Application of EPA Method 6800 in the validation of the current extraction method under study EPA Method 680038 was applied as a diagnostic tool to identify analytical biases and species transformations in the developed microwave-assisted extraction (MAE) method. SIDMS was applied as an alternative detection method to identify the steps that might alter species distribution in the MAE method protocol. Interconversions that occur after spiking are traceable and can be quantitatively corrected by

Table 7 Comparision of dierent analysis methods for the validation of the microwave-assisted extraction results. The results are expressed in mg g1 at 95% CL, n 3 HPLC-ICP-MS Sample Certied/known value/mg g1 Method 7473 (direct analysis)/mg g1 4.08 0.16 3.58 0.27 5.73 0.58 6.73 1.04 5.44 0.62 0.61 0.02 1.51 0.05 1.46 0.03 Hg21/mg g1 4.26 0.17 ND 3.02 0.06 6.06 0.56 ND 0.67 0.06 1.40 0.08 1.38 0.12 CH3Hg1/mg g1 NDa 3.81 0.20 2.66 0.07 ND 5.94 0.52 ND ND ND Total Hg/mg g1 4.26 0.17 3.81 0.20 5.68 0.09 6.06 0.56 5.94 0.52 0.67 0.06 1.40 0.08 1.38 0.12

Material-1 Inorganic mercury 4.0 Organic mercury 4.0 Mixed mercury 3.0 3.0 Material-2 Inorganic mercury 6.0 Organic mercury 6.0 Standard Reference Materials SRM 1941a 0.5 0.2 SRM 2704 1.44 0.07 SRM 2709 1.40 0.08
a

ND not detectable (lowest measurable mercury 0.5 ng ml1 and this corresponds to 5 ng g1 in soil or sediment sample).

J. Anal. At. Spectrom., 2005, 20, 183191

189

View Article Online

Table 8 Mercury species Hg21 CH3Hg1 Total Hg
a

Extraction of mercury species from BCR 580 Certied value/mg g1 NRa 0.075 0.004 132 3 DMA-80/ mg g1 NA NA 131 4 ICP-MS/ mg g1 NA NA 133 4 HPLC-ICPMS/mg g1 133 6 0.073 0.002 133 6

NR not reported. NA not applicable.

Published on 01 January 2005. Downloaded by Universiteit Utrecht on 28/08/2013 22:10:01.

Table 9 The deconvoluted concentration and transformation of mercury species in Material-1 using SIDMS calculations. Uncertainties are expressed at 95% CL with n 3 Deconvoluted concentration Hg / mg g1 DSBEa 3.05 0.12 DSAEb 2.94 0.07
a 21

Interconversion Hg21 to CH3Hg1 to CH3Hg1 (%) Hg21 (%) 1.3 1.5 0.8 1.5 0.1 1.4 0.7 0.6

CH3Hg / mg g1 2.69 0.10 2.62 0.09

b

Double spiked before extraction.

Double spiked after extraction.

no statistically signicant distinguishable interconversions were found for the materials studied; and (5) the possibility of simultaneously extracting up to ten samples, resulting in increased sample output compared with conventional extraction techniques. Since the extracts are analyzed with HPLCICP-MS for speciation, there is no need for additional steps, such as clean-up or derivatization. However, the HPLC-ICPMS technique has the limitation that it requires separation and/or preconcentration of mercury species (e.g. SCF SPE) if methylmercury is to be analyzed in sediment or soils and/or if species concentrations dier by an order of magnitude or greater. Results obtained in the analyses of two types of specically prepared reference soils and four standard reference materials (soils and sediments) containing inorganic mercury and methylmercury in four order of magnitude range veried the simplicity, eciency, precision and accuracy of the proposed microwave-assisted extraction method for mercury speciation in soils and sediments. Moreover, the application of the EPA Method 6800 as a diagnostic tool signicantly enhances the reliability of the proposed microwave-assisted extraction method. The MAE method has been chosen by the US EPA as the primary extraction protocol for mercury species extraction in EPA draft Method 3200.48

Acknowledgements
monitoring isotopes in each species.47 As SIDMS can measure the concentration of species at the time of spiking, one set of samples was double spiked before extraction and processed as described in the Experimental section. Another set of samples was extracted rst according to the MAE procedure and then double spiked and stored in a cold room until analyzed with HPLC-ICP-MS. The reference soil sample (Material-1) containing mixed mercury was used in this study for SIDMS analysis. SIDMS calculations for species transformation corrections30 were performed; the results are summarized in Table 9. It was observed that the deconvoluted concentrations for each species obtained from both sets of extractions overlap at the 95% CL and are statistically indistinguishable. Also, results for both species obtained from SIDMS calculations agree with the result obtained from method validation. Moreover, there is no statistically signicant distinguishable interconversion using the developed MAE method. Correction of conversions, when they occurred, was accomplished using EPA Method 6800 and did not alter the accuracy of the analysis. EPA describes this diagnostic and quantitative method as being a legally denitive method for such active species. Each species transformation can be tracked and corrected through this procedure. In this study, this was used to monitor and correct for specic protocol steps and was found to provide the quality assurance that was necessary to evaluate the method under study. The authors thank Science Applications International Corporation (SAIC), United States Environmental Protection Agency (US EPA) for funding and nancial support, as well as Milestone Inc., Agilent Technologies and Duquesne University for instrument and material support. Portions of this paper methodology are patented and/or patent pending.

References
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 R. P. Mason and J. M. Benoit, in Organometallic Compounds in the Environment, ed. P. J. Craig, John Wiley & Sons Ltd., Chichester, UK, 2003, p. 57. R. Rodil, A. M. Carro, R. A. Lorenzo, M. Abuin and R. Cela, J. Chromatogr., A, 2002, 963, 313323. E. Ramalhosa, S. Rio-Segade, E. Pereira, C. Vale and A. Duarte, J. Anal. At. Spectrom., 2001, 16, 643647. G. Westo o , Acta. Chem. Scand., 1967, 20, 1790. G. Westo o , Acta. Chem. Scand., 1968, 22, 2277. H. Hintelmann and R. D. Wilken, Appl. Organomet. Chem., 1993, 7, 173. P. Quevauviller, O. F. X. Donard, J. C. Wasserman, F. M. Martin and J. Schneider, Appl. Organomet. Chem., 1992, 6, 221. N. S. Bloom, Can. J. Fish. Aquat. Sci., 1989, 46, 1131. M. Horvat, N. S. Bloom and L. Liang, Anal. Chim. Acta, 1993, 281, 135. M. Horvat, V. Mandic, L. Liang, N. S. Bloom, S. Padberg, L. H. Lee, H. Hintelmann and J. Benoit, Appl. Organomet. Chem., 1994, 8, 533. M. Horvat, K. May, M. Stoeppler and A. R. Byrne, Appl. Organomet. Chem., 1988, 2, 515. Y. Han, H. M. Kingston, H. M. Boylan, G. M. M. Rahman, S. Shah, R. C. Richter, D. D. Link and S. Bhandari, Anal. Bioanal. Chem., 2003, 375(3), 428436. N. S. Bloom, J. A. Colman and L. Barber, Fresenius J. Anal. Chem., 1997, 358, 371377. H. Hintelmann, M. Hempel and R. D. Wilken, Environ. Sci. Technol., 1995, 29, 1845. J. E. Sanchez and A. Sanz-Medel, Talanta, 1998, 47, 509524. H. Emteborg, E. Bjorklund, F. Odman, L. Karlsson, L. Mathiasson, W. Frech and D. C. Baxter, Analyst, 1996, 121, 19. H. Hintelmann, R. D. Evans and J. Y. Villeneuve, J. Anal. At. Spectrom., 1995, 10(9), 619625. H. Hintelmann, R. Falter, G. Ilgen and R. D. Evans, Fresenius J. Anal. Chem., 1997, 358, 363370. H. Hintelmann and R. D. Evans, Fresenius J. Anal. Chem., 1997, 358, 378385. C. R. Hammerschmidt and W. F. Fitzgerald, Anal. Chem., 2001, 73, 59305936.

Conclusions
A simple, fast and ecient closed vessel microwave-assisted extraction method for sample preparation and mercury speciation in soils and sediments has been developed in which, after extraction with 4.0 mol l1 HNO3, inorganic and methylmercury concentrations were determined by DMA-80, ICP-MS and HPLC-ICP-MS techniques. The optimum conditions for microwave-assisted extraction of mercury species from soils and sediments were found to be 1.0 g sample, 10 ml of 4.0 mol l1 HNO3 and an irradiation time of 10 min at a temperature of 100 1C. The recoveries from the matrices analyzed were similar and quantitative. The proposed microwave-assisted extraction method oers the following advantages: (1) a notable reduction of solvent volume; (2) higher eciency of extraction achievable under optimized conditions; (3) considerable time savings in the procedure of sample preparation; (4)
190 J. Anal. At. Spectrom., 2005, 20, 183191

View Article Online

21 J. Qvarnstrom and W. Frech, J. Anal. At. Spectrom., 2002, 17, 14861491. 22 R. Falter, H. Hintelmann and P. Quevauviller, Chemosphere, 2002, 39(7), 10391049. 23 R. Falter, Chemosphere, 1999, 39(7), 10511073. 24 R. Falter, Chemosphere, 1999, 39(7), 10751091. 25 P. Quevauviller and M. Horvart, Anal. Chem., 1999, 71(5), 155A156A. 26 R. C. Rodriguez Martin-Doimeadois, M. Monperrus, E. Krupp, D. Amouroux and O.F. X. Donard, Anal. Chem., 2003, 75, 32023211. 27 H. M. Kingston, US Patent No. 5 414 259, 1995. 28 H. M. Kingston, US Patent No. 6 790 673 B1, 2004. 29 L. Lambertsson, E. Lundberg, M. Nilsson and W. Frech, J. Anal. At. Spectrom., 2001, 16, 12961301. 30 G. M. M. Rahman and H. M. S. Kingston, Anal. Chem., 2004, 76(13), 35483555. 31 X. Gaona and M. Valiente, Anal. Chim. Acta, 2003, 480, 219230. 32 B. Lind, E. Holmgren, L. Friberg and M. Vahter, Fresenius J. Anal. Chem., 1994, 348(12), 815819. 33 M. J. Vazquez, A. M. Carro, R. A. Lorenzo and R. Cela, Anal. Chem., 1997, 69, 221225. 34 M. J. Vazquez, M. Abuin, A. M. Carro, R. A. Lorenzo and R. Cela, Chemosphere, 1999, 39(7), 12111224. 35 C. M. Tseng, A. D. Diego, F. M. Martin and O. F. X. Donard, J. Anal. At. Spectrom., 1997, 12, 629635. 36 C. M. Tseng, A. D. Diego, F. M. Martin and O. F. X. Donard, J. Anal. At. Spectrom., 1997, 12, 743750. 37 C. M. Tseng, V. O. Schmitt, A. D. Diego and O. F. X. Donard, Am. Environ. Lab., 1998, 10, 2627. 38 US EPA Method 6800, Elemental and Speciated Isotope Dilution Mass Spectrometry, in Solid Waste Manual-846, Update IV, US Government Printing Oce (GPO), Washington, DC, 1998. 39 G. M. M. Rahman, H. M. S. Kingston and S. Bhandari, Appl. Organomet. Chem., 2003, 17(12), 913920. 40 J. P. Snell, C. R. Quetel, L. Lambertsson and J. Qvarnstrom, J. Anal. At. Spectrom., 2004, 19, 13151324. 41 C. Bancon-Montigny, L. Yang, R. E. Sturgeon, V. Colombini and Z. Mester, Appl. Organomet. Chem., 2004, 18, 5764. 42 R. D. Wilken and H. Hintelmann, Water, Air, Soil Pollut., 1991, 56, 427. 43 H. M. S. Kingston, P. J. Walter, W. G. Engelhart and P. J. Parsons, in Microwave Enhanced Chemistry: Fundamentals, Sample Preparation, and Applications, eds. H. M. S. Kingston and S. J. Haswell, American Chemical Society, Washington DC, 1997, pp. 697745. 44 US EPA Method 7473 in Test Method for Evaluating Solid Waste, Physical/Chemical Methods SW 846, Update IVA, US GPO, Washington DC, 2000. 45 H. M. Boylan and H. M. S. Kingston, Am. Lab., 1998, 30(19), 26. 46 G. M. M. Rahman, H. M. S. Kingston, J. C. Kern, S. Y. Yang, S. W. Hartwell and R. Anderson, Appl. Organomet. Chem., 2004, 19, in the press. 47 D. Huo, H. M. Kingston and B. Larget, in Comprehensive Analytical Chemistry, Elemental Speciation: New Approaches for Trace Element Analysis, eds. J. A. Caruso, K. L. Sutton and K. L. Ackley, Elsevier, New York, vol. 33, 2000, p. 277. 48 US EPA Draft Method 3200, Mercury Species Separation by Selective Solvent Extraction and Acid Digestion, US Government Printing Oce (GPO), Washington, DC, USA, 2004.

Published on 01 January 2005. Downloaded by Universiteit Utrecht on 28/08/2013 22:10:01.

J. Anal. At. Spectrom., 2005, 20, 183191

191